Publication Search

This study aims to develop an Indonesian traffic sign detection system using a transfer learning approach to improve road safety and traffic efficiency. The dataset was obtained from Kaggle and consists of 2,100 images across 21 traffic sign classes. The research stages include data collection, preprocessing to reduce noise and normalize image brightness, object detection using YOLOv5, and classification based on transfer learning with ResNet, VGG-16, and MobileNet architectures. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results indicate that the YOLOv5 model is capable of detecting traffic sign objects; however, the classification performance remains relatively low, with a mean Average Precision (mAP) value of 0.17. These findings suggest that further optimization is required in data preprocessing, dataset quality, and model parameter tuning to achieve better performance. This study demonstrates that transfer learning has significant potential for developing computer vision-based traffic sign detection systems, although further improvements are necessary to ensure robustness under real-world Indonesian traffic conditions.

This study aims to examine and describe the extent to which the optimization of public services in the transportation sector is implemented by the Gorontalo Regency Transportation Agency. The approach used is qualitative research with descriptive methods, in order to deeply understand the practices of transportation policies at the regional level. Data collection techniques were carried out through in-depth interviews, participatory observation, and documentation of the activities and policies implemented. The results of the study indicate that the optimization of public services in the transportation sector has been taking place progressively based on three main indicators. First, service effectiveness is reflected in the implementation of regulations in accordance with the law, such as controlling private vehicles parked haphazardly to reduce congestion and routine inspections of public vehicles to ensure safety aspects. Second, efficiency is achieved through technical supervision of vehicles that are not roadworthy, including brake conditions, tire wear, and completeness of route permits, as part of efforts to reduce operational risks and improve service quality. Third, productivity increases through the use of digital technology in socialization activities and the provision of supporting infrastructure such as bus stops, terminals, adequate roads, and traffic signs. These findings indicate that the Gorontalo Regency Transportation Agency has implemented a transportation service optimization strategy that is adaptive and oriented to community needs.

The construction of the Parepare City Twin Bridges resulted in increasing vehicle volumes and the occurrence of traffic jams which resulted in obstruction of traffic flow or congestion. The aim of this research is to determine the existing conditions, performance of intersections and road sections and their handling. The type of research used is quantitative, with the data collection method carried out (survey) and the data analysis used is MKJI 1997. The results of this research show that the existing condition of the intersection and road sections in segment 1 is at level of service (LoS) B, segment 2 v/c ratio is at level of service (LoS) A, and segment 3 v/c ratio value is at level of service (LoS) A. The performance of intersections and roads in Segment 1, namely Jl. The peak hour for the Makassar-Parepare axis is 16.00-17.00 (afternoon) Monday, the traffic flow value (Qtot) is 1812.5 pcu/hour, the saturation degree (DS) value is 0.21. Segment 2 is Jl. Gen. M. Yusuf and Jl. Arung Mampi peak hours are 16.00-17.00 (afternoon) on Mondays. The traffic flow value (Qtot) is 262.7 pcu/hour, the degree of saturation (DS) value is 0.05. The handling is installing traffic signs; installation of warning lights or traffic lights.

Educational game is a game which is specifically designed as a tool to accomplish educational needs. Therefore, it has to be projected in an exact precision to achieve its goal: educating, and increasing the user knowledge and skills. There are some problems which frequently happened at TK Kemala Bhayangkari 31 Salatiga, firstly, the number of props of traffic signs is limited so it is not proportional to the number of meetings in the existing curriculum and students must repeat the same material. Secondly, there isn’t suitable learning support media to add props to traffic signs that are still lacking. The software used in this game development was Adobe Flash CS6. The research was conducted through R&D; Research and Development method by processed through some stages. First, problem identification stage, then, data collection, product design, design validation, design revision, and the final step was the product trials. The validation was carried out by the expert validators and product user. Based on the research conducted, it concluded that this game has 43 signs which fulfill the number of signs that must be taught to students according to the curriculum at TK Kemala Bhayangkari 31 Salatiga, 18% which is most often found on the road from the total number of traffic signs in the Minister of Transportation Regulation No. 13 years 2014.  

Traffic accidents are influenced by several factors including human error, animals, weather, traffic signs. Of the four, have an influence on traffic accidents with satisfactory results. The purpose of this study was to determine whether there was an influence of Human error, Animals, Weather, and Traffic Signs on traffic accidents on the Denpasar-Gilimanuk highway, Jembrana Regency, Bali. In this study, the research population is motorized vehicle users who cross the Denpasar-Gilimanuk highway, Jembrana Regency by taking a sample system with a total of 93 respondents (samples). The variables in this study consisted of: Human error (X1), Animals (X2), Weather (X3), Traffic signs (X4) and the dependent variable was Traffic Accident (Y). The research model is multiple linear regression analysis. The results of the analysis using the statistical tool SPSS V.25 program obtained multiple linear regression equations Y = -1.530+0,313X1+0,227X2+0,258X3+0,318X4+µ. Shows that, based on empirical facts (the results of filling out the questionnaire) and multiple linear regression results show that the Human error variable (X1) has a t count of 3.680 and a t table of 1.98729 then H1 is accepted and significant, animals (X2) have a t count of 2,035 and t table of 1,98729 then H2 is accepted and significant, weather (X3) has a t count of 2.852 and t table of 1.98729 then H3 is accepted and significant, traffic signs (X4) has a t count of 4.618 and t table of 1.98729 then H4 is accepted and significant. In the coefficient of determination test (R Square) the Adjust R Square value of 0.633 means that 63.3% of the variation in the dependent variable of traffic accident factors can be explained by the independent variables of human error, animals, weather, traffic signs. While the remaining 36.7% is explained by other reasons outside the variables studied.

The purpose of this study are (1) to analyze the factors that cause violations of traffic signs by motorcycle riders in the city of Semarang. (2) To analyze criminological studies of violations of traffic signs by motorcyclists in Semarang City. The approach method is done juridically sociologically, specification descriptively using primary data and secondary data, through interviews and literature, then analyzed qualitatively. The results of the discussion show that (1) the factors that cause violations of traffic signs by motorcyclists in Semarang City are human factors, among others, the driver's indiscipline, the intent of the driver and the pretense of the driver. (2) Criminological study of violations of traffic signs by motorcycle riders in the city of Semarang, namely the presence of awareness, discipline, public interest in having a vehicle, incompetence factor